Tying a surgical knot in breast cancer outcomes

(1)

Tying a sur

gical k

not in br

east canc

er out

comes Elvir

a Lise

Vos

UITNODIGING

Voor het bijwonen van de openbare verdediging van het

proefschri�

Dinsdag 6 November 2018 Om 16:30 uur

Professor Andries Queridozaal Onderwijscentrum Erasmus MC

Wytemaweg 80 Ro� erdam Aansluitend bent u van harte uitgenodigd op de recep� e Elvira Vos Schonebergerweg 43C 3023 ZG Ro� erdam elviralvos@gmail.com Paranimfen: Githa Vos githavos@gmail.com Annemarie van Nistelrooij annemarie.avn@gmail.com

Dinsdag 6 November 2018 door

vos_cover.indd Alle pagina's 13-9-2018 20:11:36

Tying a sur

gical k

not in br

east canc

er out

comes Elvir

a Lise

Vos

UITNODIGING

Voor het bijwonen van de openbare verdediging van het

proefschri�

Dinsdag 6 November 2018 Om 16:30 uur

Professor Andries Queridozaal Onderwijscentrum Erasmus MC

Wytemaweg 80 Ro� erdam Aansluitend bent u van harte uitgenodigd op de recep� e Elvira Vos Schonebergerweg 43C 3023 ZG Ro� erdam elviralvos@gmail.com Paranimfen: Githa Vos githavos@gmail.com Annemarie van Nistelrooij annemarie.avn@gmail.com

Dinsdag 6 November 2018 door

(2)

(3)

TYING A SURGICAL KNOT

IN BREAST CANCER OUTCOMES

Het leggen van een chirurgische knoop

in de uitkomsten van borstkanker

by

Elvira Lise Vos

Vos2.indd 1 17-9-2018 11:39:04

TYING A SURGICAL KNOT

IN BREAST CANCER OUTCOMES

Het leggen van een chirurgische knoop

in de uitkomsten van borstkanker

by

Elvira Lise Vos

Vos2.indd 1 17-9-2018 11:39:04

(4)

The studies described in this thesis were supported by Stichting Theia – Zilveren Kruis and the Dutch Cancer Society [grant nr EMCR 2015-7784].

Printing of this thesis was financially supported by the Dutch Cancer Society, the Netherlands Comprehensive Cancer Organisation and ChipSoft.

Cover design and layout by: Remco Wetzels Printing by: Ridderprint

Copyright @ E.L. Vos, Rotterdam, the Netherlands

No parts of this thesis may be reproduced, stored in a retrieval system, or trans-mitted in any form or by any means without prior permission from the author or copyright-owning journals for previously published chapters.

Vos2.indd 2 17-9-2018 11:39:04

The studies described in this thesis were supported by Stichting Theia – Zilveren Kruis and the Dutch Cancer Society [grant nr EMCR 2015-7784].

Printing of this thesis was financially supported by the Dutch Cancer Society, the Netherlands Comprehensive Cancer Organisation and ChipSoft.

Cover design and layout by: Remco Wetzels Printing by: Ridderprint

Copyright @ E.L. Vos, Rotterdam, the Netherlands

No parts of this thesis may be reproduced, stored in a retrieval system, or trans-mitted in any form or by any means without prior permission from the author or copyright-owning journals for previously published chapters.

Vos2.indd 2 17-9-2018 11:39:04

(5)

TYING A SURGICAL KNOT

IN BREAST CANCER OUTCOMES

Het leggen van een chirurgische knoop

in de uitkomsten van borstkanker

Thesis

to obtain the degree of Doctor from the

Erasmus University Rotterdam

by command of the

rector magnificus

Prof.dr. R.C.M.E. Engels

and in accordance with the decision of the Doctorate Board.

The public defense shall be held on

Tuesday 6 November 2018 at 15:30hrs

by

Elvira Lise Vos

born in Ridderkerk

Vos2.indd 3 17-9-2018 11:39:04

TYING A SURGICAL KNOT

IN BREAST CANCER OUTCOMES

Het leggen van een chirurgische knoop

in de uitkomsten van borstkanker

Thesis

to obtain the degree of Doctor from the

Erasmus University Rotterdam

by command of the

rector magnificus

Prof.dr. R.C.M.E. Engels

and in accordance with the decision of the Doctorate Board.

The public defense shall be held on

Tuesday 6 November 2018 at 15:30hrs

by

Elvira Lise Vos

born in Ridderkerk

Vos2.indd 3 17-9-2018 11:39:04

(6)

DOCTORAL COMMITTEE

Promotors:

Prof.dr. C. Verhoef

Prof.dr.

S. Siesling

Other members:

Prof.dr. M.G.M. Hunink

Prof.dr.

E.J.Th.

Rutgers

Prof.dr.

J.P.

Pignol

Co-promotor:

Dr. L.B. Koppert

Vos2.indd 4 17-9-2018 11:39:04

DOCTORAL COMMITTEE

Promotors:

Prof.dr. C. Verhoef

Prof.dr.

S. Siesling

Other members:

Prof.dr. M.G.M. Hunink

Prof.dr.

E.J.Th.

Rutgers

Prof.dr.

J.P.

Pignol

Co-promotor:

Dr. L.B. Koppert

Vos2.indd 4 17-9-2018 11:39:04

(7)

To my parents

Vos2.indd 7 17-9-2018 11:39:04

To my parents

Vos2.indd 7 17-9-2018 11:39:04

(10)

Vos2.indd 8 17-9-2018 11:39:05 Vos2.indd 8 17-9-2018 11:39:05

(11)

PART I

Introducti on

Vos2.indd 9 17-9-2018 11:39:06

PART I

Introducti on

Vos2.indd 9 17-9-2018 11:39:06

(12)

Vos2.indd 10 17-9-2018 11:39:06 Vos2.indd 10 17-9-2018 11:39:06

(13)

CHAPTER 1 General introducti on and

outline of this thesis

Vos2.indd 11 17-9-2018 11:39:08

CHAPTER 1 General introducti on and

outline of this thesis

Vos2.indd 11 17-9-2018 11:39:08

(14)

INTRODUCTION

The present thesis focuses on outcomes of breast conserving surgery in early stage breast cancer patients. These outcomes include cosmetic results, quality of life, radicality, reexcision, secondary mastectomy, local recurrence, overall survival and quality of care. This thesis provides insight in how these outcomes are intertwined. To explain the title of this thesis, tying a surgical knot is a metaphor for intertwining. The different breast cancer surgery outcomes represent the surgical sutures. Breast cancer

Anatomy of the Breast

The breast or so called mammary gland contains skin, adipose tissue, and fibroglandular tissue. Beneath the skin lays the superficial fascia which contains the fibroglandular tissue or so called breast parenchyma. The deep fascia lies anterior to mainly the pectoralis major muscle. Fibrous connective tissue connects the superficial and deep fascia and inserts into the dermis. The breast parenchyma contains 15-20 lobes that are made up of smaller lobules and is connected with the nipple by ducts. There is a network of blood vessels, nerves, lymph vessels, and lymph nodes. The shape, contour, volume, and density of the breasts varies significantly between individuals. For practical reasons, the breast can be divided into four quadrants: upper medial, upper lateral, lower lateral, and lower medial. The upper lateral quadrant usually contain more fibroglandular tissue and therefore is the most frequent location of breast cancer.

Diagnosis

Breast Cancer is the most common type of cancer in woman in the Netherlands. About 1 in 8 woman are diagnosed with breast cancer during her lifetime. In 2016, 14,451 new cases of invasive breast cancer and 2,592 new cases of ductal carcinoma-in-situ (DCIS) were registered by the Netherlands Cancer Registry(1). One-hundred thirty-six cases concerned males. DCIS is a precursor lesion of invasive breast cancer and is therefore also called non-invasive breast cancer. It is found in the ducts of the breast and has not developed the ability to invade outside the ducts into surrounding breast tissue. Invasive breast cancer is most commonly of the ductal type, meaning it arose in the ducts, followed by the lobular type, meaning it arose in the lobules. Other types of breast cancer exist, but are uncommon and not addressed in current thesis. In 80% of the breast cancer patients it is diagnosed at an early stage, i.e., stage I/II as defined by the tumor node metastasis staging (TNM) Classification of Malignant Tumors by the American

Vos2.indd 12 17-9-2018 11:39:08

INTRODUCTION

The present thesis focuses on outcomes of breast conserving surgery in early stage breast cancer patients. These outcomes include cosmetic results, quality of life, radicality, reexcision, secondary mastectomy, local recurrence, overall survival and quality of care. This thesis provides insight in how these outcomes are intertwined. To explain the title of this thesis, tying a surgical knot is a metaphor for intertwining. The different breast cancer surgery outcomes represent the surgical sutures. Breast cancer

Anatomy of the Breast

The breast or so called mammary gland contains skin, adipose tissue, and fibroglandular tissue. Beneath the skin lays the superficial fascia which contains the fibroglandular tissue or so called breast parenchyma. The deep fascia lies anterior to mainly the pectoralis major muscle. Fibrous connective tissue connects the superficial and deep fascia and inserts into the dermis. The breast parenchyma contains 15-20 lobes that are made up of smaller lobules and is connected with the nipple by ducts. There is a network of blood vessels, nerves, lymph vessels, and lymph nodes. The shape, contour, volume, and density of the breasts varies significantly between individuals. For practical reasons, the breast can be divided into four quadrants: upper medial, upper lateral, lower lateral, and lower medial. The upper lateral quadrant usually contain more fibroglandular tissue and therefore is the most frequent location of breast cancer.

Diagnosis

Breast Cancer is the most common type of cancer in woman in the Netherlands. About 1 in 8 woman are diagnosed with breast cancer during her lifetime. In 2016, 14,451 new cases of invasive breast cancer and 2,592 new cases of ductal carcinoma-in-situ (DCIS) were registered by the Netherlands Cancer Registry(1). One-hundred thirty-six cases concerned males. DCIS is a precursor lesion of invasive breast cancer and is therefore also called non-invasive breast cancer. It is found in the ducts of the breast and has not developed the ability to invade outside the ducts into surrounding breast tissue. Invasive breast cancer is most commonly of the ductal type, meaning it arose in the ducts, followed by the lobular type, meaning it arose in the lobules. Other types of breast cancer exist, but are uncommon and not addressed in current thesis. In 80% of the breast cancer patients it is diagnosed at an early stage, i.e., stage I/II as defined by the tumor node metastasis staging (TNM) Classification of Malignant Tumors by the American

Vos2.indd 12 17-9-2018 11:39:08

(15)

GENERAL INTRODUCTION AND OUTLINE OF THIS THESIS |

1

13

1

Joint Committee on Cancer(2). Imaging consists of mammography and ultrasound

in all patients and magnetic resonance imaging (MRI) in some patients. MRI increases the detection of additional disease, but studies also suggest it leads to more mastectomies(3,4). Guidelines give varying recommendations regarding the use of MRI and this is reflected in wide interhospital variation(5). Other imaging modalities are not addressed in current thesis.

Surgical treatment

Surgery remains the primary treatment composing of mastectomy or breast conserving surgery (BCS). Randomized phase III trials have proven that survival is equivalent for mastectomy and BCS followed by radiation therapy for stage I/II patients after 20-years of follow-up even though that local recurrence rates may differ(6-8). Recent retrospective analysis of large population-based data have shown that BCS with radiation therapy may even be preferred over mastectomy regarding survival(9-12). Mastectomy can be followed by immediate or delayed implant-based or autologous breast reconstruction resulting in similar cosmetic satisfaction as compared to breast conservation(13).

Cosmetic outcome and quality of life

Cosmetic outcome influences quality of life (QoL) and plays an important role in the treatment decision(14, 15). BCS should only be offered if no contra-indication exists and if an acceptable cosmetic outcome can be achieved. However judging the feasibility of attaining favorable cosmetic result by BCS is extremely subjective and can be rather difficult. The subjectivity of the treatment decision is reflected in the considerable variation of BCS rates between hospitals(16). Furthermore, no golden standard for measuring cosmetic outcome exists raising a barrier for taking cosmetic outcome into consideration. Larger excision volumes have been correlated to unfavorable cosmetic outcome(17-19). This is important when taking into consideration that complete tumor excision (i.e., radicality) is another outcome of BCS. A larger excision facilitates achieving radicality and a narrow excision benefits cosmetic outcome. However, if the excision is too narrow and turns out to be irradical, then a reexcision may be necessary which again worsens cosmetic result(19). Here is an emerging role for direct oncoplastic reconstruction. That means applying plastic surgical techniques to the BCS attempting to conserve the breast with favorable cosmetic outcome(20, 21). Thereby it can prevent mastectomy or cosmetic deformity following BCS while allowing larger excisions. The indications for direct oncoplastic reconstructions are yet unclear. Another

Vos2.indd 13 17-9-2018 11:39:08

1

13

1

Joint Committee on Cancer(2). Imaging consists of mammography and ultrasound

in all patients and magnetic resonance imaging (MRI) in some patients. MRI increases the detection of additional disease, but studies also suggest it leads to more mastectomies(3,4). Guidelines give varying recommendations regarding the use of MRI and this is reflected in wide interhospital variation(5). Other imaging modalities are not addressed in current thesis.

Surgical treatment

Surgery remains the primary treatment composing of mastectomy or breast conserving surgery (BCS). Randomized phase III trials have proven that survival is equivalent for mastectomy and BCS followed by radiation therapy for stage I/II patients after 20-years of follow-up even though that local recurrence rates may differ(6-8). Recent retrospective analysis of large population-based data have shown that BCS with radiation therapy may even be preferred over mastectomy regarding survival(9-12). Mastectomy can be followed by immediate or delayed implant-based or autologous breast reconstruction resulting in similar cosmetic satisfaction as compared to breast conservation(13).

Cosmetic outcome and quality of life

Cosmetic outcome influences quality of life (QoL) and plays an important role in the treatment decision(14, 15). BCS should only be offered if no contra-indication exists and if an acceptable cosmetic outcome can be achieved. However judging the feasibility of attaining favorable cosmetic result by BCS is extremely subjective and can be rather difficult. The subjectivity of the treatment decision is reflected in the considerable variation of BCS rates between hospitals(16). Furthermore, no golden standard for measuring cosmetic outcome exists raising a barrier for taking cosmetic outcome into consideration. Larger excision volumes have been correlated to unfavorable cosmetic outcome(17-19). This is important when taking into consideration that complete tumor excision (i.e., radicality) is another outcome of BCS. A larger excision facilitates achieving radicality and a narrow excision benefits cosmetic outcome. However, if the excision is too narrow and turns out to be irradical, then a reexcision may be necessary which again worsens cosmetic result(19). Here is an emerging role for direct oncoplastic reconstruction. That means applying plastic surgical techniques to the BCS attempting to conserve the breast with favorable cosmetic outcome(20, 21). Thereby it can prevent mastectomy or cosmetic deformity following BCS while allowing larger excisions. The indications for direct oncoplastic reconstructions are yet unclear. Another

Vos2.indd 13 17-9-2018 11:39:08

(16)

| CHAPTER 1 14

treatment strategy is neo-adjuvant chemotherapy that can reduce tumor size and increases likelihood of achieving breast conservation with favorable cosmetic outcome. Cosmetic outcome is also influenced by the radiation therapy following BCS. Moderate to severe fibrosis occurs in 26% of irradiated patients(22). Sometimes even a radiation boost dose is added since it has proven to reduce local recurrence risk in some patients, another important outcome after BCS(23). In turn however it increases breast fibrosis and worsens cosmetic outcome(24).

Treatment decision making based on quality of life

Quality-adjusted life years (QALYs) are commonly used outcome measures for comparing effectiveness of treatments. QALYs provide a metric for valuing the impact of treatment on health-related quality of life on a common scale(25). This is achieved by signing a utility value to each health state on a scale from 0 (i.e., dead) to 1 (i.e., full health). A utility is a quantitative measure of the strength of a person’s preference for an outcome(26). There are many ways to derive these so-called health state utility values (HSUVs). Commonly used HSUVs are generic preference-based derived by the EuroQol Group with the EQ-5D instrument. This questionnaire has five dimensions each with three levels and thus defines 243 health states. Each state has a value on the 0-1 scale obtained by interviewing a sample of the general population. When the EQ-5D questionnaire is completed by patients it results in a utility value for the patients’ health state. In contrast to a condition specific preference-based measure, it may not be sensitive to show quality of life differences between all patient groups.

A treatment decision model can calculate the treatment threshold when to treat for optimal QALYs(26). A decision tree models these different treatment consequences based on certain probabilities. Each endpoint in the decision tree represents a health state to which a utility value is attached. The definition of a treatment threshold is the probability of disease at which the expected value of treatment and no treatment are equal. The threshold can be determined by direct comparison of the benefits and harms of treatment(26).

Oncological outcomes

Radicality, reexcision and secondary mastectomy

As one important outcome of the surgical procedure, the surgical margins of the excised specimen are examined directly postoperatively by the pathologist. The margins can be defined as negative (i.e., radical or ‘no tumor touching the inked margins’) or positive (i.e., irradical or ‘tumor touching the inked margins’).

Vos2.indd 14 17-9-2018 11:39:08

| CHAPTER 1 14

treatment strategy is neo-adjuvant chemotherapy that can reduce tumor size and increases likelihood of achieving breast conservation with favorable cosmetic outcome. Cosmetic outcome is also influenced by the radiation therapy following BCS. Moderate to severe fibrosis occurs in 26% of irradiated patients(22). Sometimes even a radiation boost dose is added since it has proven to reduce local recurrence risk in some patients, another important outcome after BCS(23). In turn however it increases breast fibrosis and worsens cosmetic outcome(24).

Treatment decision making based on quality of life

Quality-adjusted life years (QALYs) are commonly used outcome measures for comparing effectiveness of treatments. QALYs provide a metric for valuing the impact of treatment on health-related quality of life on a common scale(25). This is achieved by signing a utility value to each health state on a scale from 0 (i.e., dead) to 1 (i.e., full health). A utility is a quantitative measure of the strength of a person’s preference for an outcome(26). There are many ways to derive these so-called health state utility values (HSUVs). Commonly used HSUVs are generic preference-based derived by the EuroQol Group with the EQ-5D instrument. This questionnaire has five dimensions each with three levels and thus defines 243 health states. Each state has a value on the 0-1 scale obtained by interviewing a sample of the general population. When the EQ-5D questionnaire is completed by patients it results in a utility value for the patients’ health state. In contrast to a condition specific preference-based measure, it may not be sensitive to show quality of life differences between all patient groups.

A treatment decision model can calculate the treatment threshold when to treat for optimal QALYs(26). A decision tree models these different treatment consequences based on certain probabilities. Each endpoint in the decision tree represents a health state to which a utility value is attached. The definition of a treatment threshold is the probability of disease at which the expected value of treatment and no treatment are equal. The threshold can be determined by direct comparison of the benefits and harms of treatment(26).

Oncological outcomes

Radicality, reexcision and secondary mastectomy

As one important outcome of the surgical procedure, the surgical margins of the excised specimen are examined directly postoperatively by the pathologist. The margins can be defined as negative (i.e., radical or ‘no tumor touching the inked margins’) or positive (i.e., irradical or ‘tumor touching the inked margins’).

Vos2.indd 14 17-9-2018 11:39:08

(17)

1

15

1

If the margins are positive, the excision is incomplete and residual disease may

be left behind in the breast. Worldwide there was no consensus on how wide the cancer-free margin should be until 2014. This controversy was reflected by the large variation of reexcision indications used between hospitals and clinicians from a 10mm negative margin to a tumor positive margin(27). High level evidence about an acceptable margin for BCS is lacking. From the large randomized trials between mastectomy and BCS with radiation therapy, only the NSABP B-06 trial required tumor free resection margins meaning that other trials included patients with incomplete resection margins(6-8). A meta-analysis of retrospective studies showed that positive resection margins independently increased the risk on local recurrence as compared to negative margins(28). This resulted in the recommendation to perform reexcision for positive margins by the Society of Surgical Oncology-American Society for Radiation Oncology (SSO/ASTRO) in 2014(29).

The Dutch national guideline distinguishes between a focally positive margin (i.e., invasive tumor and/or DCIS touching the inked margin over a length of 4mm or less) and an extensively positive margin (i.e., tumor touching the inked margin over a length of more than 4mm). The Dutch guideline already since 2002 does not recommend a reexcision for focally positive margins, but does recommends to apply whole-breast radiation therapy including a radiation boost dose(30). Currently, the choice of treatment (i.e., reexcision or not and radiation boost or not) is provider dependent in the Netherlands. However reexcisions increase healthcare costs, increase burden to the patients, worsen cosmetic outcome and quality of life, and result in secondary mastectomies.

Local recurrence and overall survival

Prognosis, measured in outcomes such as local recurrence and overall survival, has improved due to changes in the multidisciplinary treatment of breast cancer. Patients treated in the 1980s had a 10-year local recurrence rate of 10-20%(31, 32) after BCS. With the introduction of radiation therapy boost, early stage breast cancer patients treated with BCS in the 1990s had a 10-year local recurrence rate of 6.2%(22). As explained earlier, at the cost of cosmetic outcome. Patients treated in the 2000s had a 5-year local recurrence rate of 2.7%(33). The 10-year overall survival rate of patients treated with BCS in 1980s was 65%(32), in the 1990s was 82%(22) and further improved in the 2000s(9, 34). Patients treated between 2006-2012 had an even significant better overall survival as compared to patients treated between 1999-2005 before important changes in the indication and type of systemic therapy in daily practice occurred(35).

Vos2.indd 15 17-9-2018 11:39:08

1

15

1

If the margins are positive, the excision is incomplete and residual disease may

be left behind in the breast. Worldwide there was no consensus on how wide the cancer-free margin should be until 2014. This controversy was reflected by the large variation of reexcision indications used between hospitals and clinicians from a 10mm negative margin to a tumor positive margin(27). High level evidence about an acceptable margin for BCS is lacking. From the large randomized trials between mastectomy and BCS with radiation therapy, only the NSABP B-06 trial required tumor free resection margins meaning that other trials included patients with incomplete resection margins(6-8). A meta-analysis of retrospective studies showed that positive resection margins independently increased the risk on local recurrence as compared to negative margins(28). This resulted in the recommendation to perform reexcision for positive margins by the Society of Surgical Oncology-American Society for Radiation Oncology (SSO/ASTRO) in 2014(29).

The Dutch national guideline distinguishes between a focally positive margin (i.e., invasive tumor and/or DCIS touching the inked margin over a length of 4mm or less) and an extensively positive margin (i.e., tumor touching the inked margin over a length of more than 4mm). The Dutch guideline already since 2002 does not recommend a reexcision for focally positive margins, but does recommends to apply whole-breast radiation therapy including a radiation boost dose(30). Currently, the choice of treatment (i.e., reexcision or not and radiation boost or not) is provider dependent in the Netherlands. However reexcisions increase healthcare costs, increase burden to the patients, worsen cosmetic outcome and quality of life, and result in secondary mastectomies.

Local recurrence and overall survival

Prognosis, measured in outcomes such as local recurrence and overall survival, has improved due to changes in the multidisciplinary treatment of breast cancer. Patients treated in the 1980s had a 10-year local recurrence rate of 10-20%(31, 32) after BCS. With the introduction of radiation therapy boost, early stage breast cancer patients treated with BCS in the 1990s had a 10-year local recurrence rate of 6.2%(22). As explained earlier, at the cost of cosmetic outcome. Patients treated in the 2000s had a 5-year local recurrence rate of 2.7%(33). The 10-year overall survival rate of patients treated with BCS in 1980s was 65%(32), in the 1990s was 82%(22) and further improved in the 2000s(9, 34). Patients treated between 2006-2012 had an even significant better overall survival as compared to patients treated between 1999-2005 before important changes in the indication and type of systemic therapy in daily practice occurred(35).

Vos2.indd 15 17-9-2018 11:39:08

(18)

| CHAPTER 1 16

Quality of care

Quality of care is an outcome measure generally assessed by quality indicators and classified into structure, process, and outcome type(36). Structure indicators define the characteristics of the hospital in which the care is provided, e.g. hospital patient volume. Process indicators refer to the appropriateness of the delivered care, e.g. proportion of patients who received a full pathology report. Outcome indicators reflect the result as a consequence of care, e.g. proportion of patients with a reexcision. Quality of care information can be used for internal and exter-nal purposes. Exterexter-nal use involves the public comparison of individual hospitals. Comparison to the average is called benchmarking and the construction of league tables is called ranking(37). Public reporting of quality indicator outcomes stimu-lates quality improvement activities by hospitals(38). Moreover, the demand for quality assurance and transparency is rapidly increasing worldwide from govern-ment agencies, accrediting bodies, medical specialty societies, health care insur-ance companies, and patient organizations. To prevent misinterpretation of data, it is important that quality indicator differences between hospitals represent real dif-ferences in quality of care. Therefore the scientific rigor of quality indicators should be evaluated, i.e., the validity and reliability(39). Reliability refers to the degree to which an indicator is reproducible. It is influenced by data quality, accuracy of indi-cator definitions, and statistical uncertainty caused by low number of events(40). Validity refers to whether an indicator measures what it claims to measure. Usually quality indicators are defined based on expert opinion and the available evidence by face validity(41). Face validity means subjectively a quality indicator measures what it is intended to measure. Construct validity refers to evaluating the relation between indicators that measure the same underlying concept(42). Validity is in-fluenced by systematic errors, e.g. baseline risk differences between patient popu-lations (i.e., case-mix)(43). A part of the observed differences between hospitals is often explained by statistical uncertainty and case-mix that can be measured. The remaining part of observed differences between hospitals represent unexplained differences. These might be due to the quality of care. The percentage of unex-plained differences is called rankability and refers to the reliability of ranking hos-pitals based on quality indicator information(44).

Vos2.indd 16 17-9-2018 11:39:08

| CHAPTER 1 16

Quality of care

Quality of care is an outcome measure generally assessed by quality indicators and classified into structure, process, and outcome type(36). Structure indicators define the characteristics of the hospital in which the care is provided, e.g. hospital patient volume. Process indicators refer to the appropriateness of the delivered care, e.g. proportion of patients who received a full pathology report. Outcome indicators reflect the result as a consequence of care, e.g. proportion of patients with a reexcision. Quality of care information can be used for internal and exter-nal purposes. Exterexter-nal use involves the public comparison of individual hospitals. Comparison to the average is called benchmarking and the construction of league tables is called ranking(37). Public reporting of quality indicator outcomes stimu-lates quality improvement activities by hospitals(38). Moreover, the demand for quality assurance and transparency is rapidly increasing worldwide from govern-ment agencies, accrediting bodies, medical specialty societies, health care insur-ance companies, and patient organizations. To prevent misinterpretation of data, it is important that quality indicator differences between hospitals represent real dif-ferences in quality of care. Therefore the scientific rigor of quality indicators should be evaluated, i.e., the validity and reliability(39). Reliability refers to the degree to which an indicator is reproducible. It is influenced by data quality, accuracy of indi-cator definitions, and statistical uncertainty caused by low number of events(40). Validity refers to whether an indicator measures what it claims to measure. Usually quality indicators are defined based on expert opinion and the available evidence by face validity(41). Face validity means subjectively a quality indicator measures what it is intended to measure. Construct validity refers to evaluating the relation between indicators that measure the same underlying concept(42). Validity is in-fluenced by systematic errors, e.g. baseline risk differences between patient popu-lations (i.e., case-mix)(43). A part of the observed differences between hospitals is often explained by statistical uncertainty and case-mix that can be measured. The remaining part of observed differences between hospitals represent unexplained differences. These might be due to the quality of care. The percentage of unex-plained differences is called rankability and refers to the reliability of ranking hos-pitals based on quality indicator information(44).

Vos2.indd 16 17-9-2018 11:39:08

(19)

1

17

1 AIMS AND OUTLINE

The main aim of this thesis is to add knowledge on how to improve outcomes of breast cancer surgery.

Part II investigates the value of a new objective tool to predict cosmetic outcome after BCS in early stage breast cancer patients in Chapter 2. How this prediction can be used as a treatment decision model between BCS and mastectomy for optimal QoL is studied in Chapter 3. Then a randomized controlled trial is proposed

in Chapter 4. It aims is to improve cosmetic outcome and QoL by use of the

treatment decision model in clinical practice.

Part III starts with focusing on radicality, reexcision, and secondary mastectomy after BCS. Chapter 5 aims to identify subgroups in which the preoperative use of MRI might improve these outcomes. Then it focuses on focally positive margins and the need for a reexcision. Clinicopathological factors associated with focally positive margins and the presence of residual disease are identified in Chapter 6. The incidence of residual disease according to negative margins, focally positive margins, and extensively positive margins is studied. Chapter 7 shines a light on the safety of the Dutch guideline recommending to perform reexcision in case of extensively positive margins only. It compares overall survival in patients with a reexcision after BCS to patients without a reexcision. Chapter 8 describes how often the Dutch recommendation is followed in clinical practice and reexcision is omitted in patients with focally positive margins after BCS. It then studies if local recurrence rate, disease free survival, and overall survival is impaired in this specific patient group.

Part IV studies quality indicators for breast cancer care that are used in the Netherlands. Chapter 9 quantifies the influence of case-mix and statistical uncertainty and the remaining between hospital differences that may be due to true differences in quality of care. Construct validity between indicators is tested

in Chapter 10. Quality indicators are then selected for a new case-mix corrected

summarizing measure and hospitals are ranked.

Part V includes a summary in Chapter 11. The results from previous chapters are discussed and future perspectives are outlined in Chapter 12.

Vos2.indd 17 17-9-2018 11:39:08

1

17

1 AIMS AND OUTLINE

The main aim of this thesis is to add knowledge on how to improve outcomes of breast cancer surgery.

Part II investigates the value of a new objective tool to predict cosmetic outcome after BCS in early stage breast cancer patients in Chapter 2. How this prediction can be used as a treatment decision model between BCS and mastectomy for optimal QoL is studied in Chapter 3. Then a randomized controlled trial is proposed

in Chapter 4. It aims is to improve cosmetic outcome and QoL by use of the

treatment decision model in clinical practice.

Part III starts with focusing on radicality, reexcision, and secondary mastectomy after BCS. Chapter 5 aims to identify subgroups in which the preoperative use of MRI might improve these outcomes. Then it focuses on focally positive margins and the need for a reexcision. Clinicopathological factors associated with focally positive margins and the presence of residual disease are identified in Chapter 6. The incidence of residual disease according to negative margins, focally positive margins, and extensively positive margins is studied. Chapter 7 shines a light on the safety of the Dutch guideline recommending to perform reexcision in case of extensively positive margins only. It compares overall survival in patients with a reexcision after BCS to patients without a reexcision. Chapter 8 describes how often the Dutch recommendation is followed in clinical practice and reexcision is omitted in patients with focally positive margins after BCS. It then studies if local recurrence rate, disease free survival, and overall survival is impaired in this specific patient group.

Part IV studies quality indicators for breast cancer care that are used in the Netherlands. Chapter 9 quantifies the influence of case-mix and statistical uncertainty and the remaining between hospital differences that may be due to true differences in quality of care. Construct validity between indicators is tested

in Chapter 10. Quality indicators are then selected for a new case-mix corrected

summarizing measure and hospitals are ranked.

Part V includes a summary in Chapter 11. The results from previous chapters are discussed and future perspectives are outlined in Chapter 12.

Vos2.indd 17 17-9-2018 11:39:08

(20)

| CHAPTER 1 18

REFERENCES

1. Nederlandse Kankerregistratie. beheerd door IKNL. March 2018. 2. (AJCC) AJCoC. AJCC Cancer Staging Manual.

3. Houssami N, Ciatto S, Macaskill P, Lord SJ, Warren RM, Dixon JM, et al. Accuracy and surgical im-pact of magnetic resonance imaging in breast cancer staging: systematic review and meta-anal-ysis in detection of multifocal and multicentric cancer. J Clin Oncol. 2008;26(19):3248-58. 4. Kuhl CK, Schrading S, Bieling HB, Wardelmann E, Leutner CC, Koenig R, et al. MRI for diagnosis of

pure ductal carcinoma in situ: a prospective observational study. Lancet. 2007;370(9586):485-92.

5. Dutch Institute for Clinical Auditing (DICA) / NABON Breast Cancer Audit (NBCA). http://nbcaclin-icalauditnl/.

6. Veronesi U, Cascinelli N, Mariani L, Greco M, Saccozzi R, Luini A, et al. Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med. 2002;347(16):1227-32.

7. Fisher B, Anderson S, Bryant J, Margolese RG, Deutsch M, Fisher ER, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradia-tion for the treatment of invasive breast cancer. N Engl J Med. 2002;347(16):1233-41.

8. Litiere S, Werutsky G, Fentiman IS, Rutgers E, Christiaens MR, Van Limbergen E, et al. Breast con-serving therapy versus mastectomy for stage I-II breast cancer: 20 year follow-up of the EORTC 10801 phase 3 randomised trial. Lancet Oncol. 2012;13(4):412-9.

9. Lagendijk M, van Maaren MC, Saadatmand S, Strobbe LJA, Poortmans PMP, Koppert LB, et al. Breast conserving therapy and mastectomy revisited: Breast cancer-specific survival and the in-fluence of prognostic factors in 129,692 patients. Int J Cancer. 2018;142(1):165-75.

10. van Maaren MC, de Munck L, de Bock GH, Jobsen JJ, van Dalen T, Linn SC, et al. 10 year survival after breast-conserving surgery plus radiotherapy compared with mastectomy in early breast cancer in the Netherlands: a population-based study. Lancet Oncol. 2016;17(8):1158-70. 11. Hartmann-Johnsen OJ, Karesen R, Schlichting E, Nygard JF. Survival is Better After Breast

Con-serving Therapy than Mastectomy for Early Stage Breast Cancer: A Registry-Based Follow-up Study of Norwegian Women Primary Operated Between 1998 and 2008. Ann Surg Oncol. 2015;22(12):3836-45.

12. Agarwal S, Pappas L, Neumayer L, Kokeny K, Agarwal J. Effect of breast conservation ther-apy vs mastectomy on disease-specific survival for early-stage breast cancer. JAMA Surg. 2014;149(3):267-74.

13. Jagsi R, Li Y, Morrow M, Janz N, Alderman A, Graff J, et al. Patient-Reported Quality of Life and Sat-isfaction With Cosmetic Outcomes After Breast Conservation and Mastectomy With and With-out Reconstruction: Results of a Survey of Breast Cancer Survivors. Ann Surg. 2015;261(6):1198-206.

14. Hau E, Browne L, Capp A, Delaney GP, Fox C, Kearsley JH, et al. The impact of breast cosmetic and functional outcomes on quality of life: long-term results from the St. George and Wollongong randomized breast boost trial. Breast Cancer Res Treat. 2013;139(1):115-23.

15. Dahlback C, Ullmark JH, Rehn M, Ringberg A, Manjer J. Aesthetic result after breast-conserving therapy is associated with quality of life several years after treatment. Swedish women evaluat-ed with BCCT.core and BREAST-Q. Breast Cancer Res Treat. 2017;164(3):679-87.

Vos2.indd 18 17-9-2018 11:39:08